High-Performance Algorithms and Complex Fluids
NREL tackles problems at the limits of what is computationally possible, working to formulate and implement algorithms that help solve the underlying partial differential equations on upcoming computational architectures.
Using high-performance computing (HPC) to predict and understand the behavior of complex fluid mechanics is a foundational activity in the pursuit of basic and early-stage applied research motivated by reacting, multiphase, and traditional internal and external flows. Such flows arise in many areas such as vehicle or power generation combustion and in wind farms. NREL also pursues data assimilation techniques: parameter estimation, optimization, machine learning, and surrogate model construction to develop insights only possible by combining experimental data with simulation.
- Block structure adaptive mesh refinement, overset and sliding meshes, unstructured grids, cartesian cut-cell, and multi-physics simulations
- Exascale computing software development and HPC performance optimization
- The mathematics of combining data and simulation
- Computational fluid mechanics simulations of reacting, particle-laden, non-Newtonian, as well as traditional internal and external flows.